diff --git a/kats/models/globalmodel/utils.py b/kats/models/globalmodel/utils.py index 7b7d992e2..16181bb7c 100644 --- a/kats/models/globalmodel/utils.py +++ b/kats/models/globalmodel/utils.py @@ -1153,7 +1153,7 @@ class GMParam: validation_metric: Optional; A list of strings representing the names of the error metrics for validation. Default is None, which sets validation_metric as ['smape', 'sbias', 'exceed']. batch_size: Optional; A dictionary representing the batch size schedule, whose keys are the epoch numbers and the corresponding values are the batch sizes. Default is None, which sets batch_size as {0:2,3:5,4:15,5:50,6:150,7:500}. learning_rate: Optional; A dictionary representing the learning rate schedule, whose keys are epoch numbers and the corresponding values are the learning rates. Default is None, which sets learning_rate as {0: 1e-3, 2: 1e-3/3.}. - epoch_num: Optional; An integer representing the totoal number of epoches. Default is 8. + epoch_num: Optional; An integer representing the total number of epoches. Default is 8. epoch_size: Optional; An integer representing the batches per epoch. Default is 3000. init_seasonality: Optional; A list of two floats representing the lower and upper bounds for the initial seasonalities. Default is None, which sets init_seasonality as [0.1, 10.]. init_smoothing_params: Optional; A list of two floats representing initial values for smoothing parameters, i.e., level smoothing parameter and seasonality smoothing parameter. Default is None, which sets init_smoothing_params as [0.4, 0.6].